The acquisition and recall of learned information is modulated by the circadian system, and disruption of circadian timing by jet-lag or shift-work impairs memory. Circadian timing originates in the suprachiasmatic nucleus (SCN), and the hippocampus is well understood for its role in certain types of memory (e.g., spatial, contextual). The SCN does not innervate the hippocampus directly, but innervates the medial septum (MS), which is the primary subcortical input to the hippocampus. Over 20 years ago, this pathway was proposed as the mechanism by which the SCN could modulate learning and memory (Watts, 1991), but this idea has never been tested. The lack of research on this topic is remarkable considering that one of the major problems in the field of circadian rhythms has been to identify functional output pathways of the SCN. Circadian timing is easily eliminated in Siberian hamsters (Phodopus sungorus) by exposing them to a phase-advancing and a phase-delaying light signal given within one circadian cycle. This arrhythmia results in major memory deficits in object recognition and spatial navigation. These deficits were not due to motivational, perceptual, or attentional factors, nor were they due to sleep-related issues. We propose that the arrhythmic SCN releases GABA (its principal neurotransmitter) in a noncircadian manner, and thereby provides continuous inhibition of the MS. There is a robust literature showing that the GABAA agonist, muscimol, reduces synaptic excitability in the hippocampus and impairs memory. Our studies show that systemic injections of the GABAA receptor antagonist, pentylenetetrazole (PTZ) completely restore object recognition and spatial memory in arrhythmic hamsters. Therefore, we propose to test the hypothesis that chronic SCN GABAergic inhibition of the MS in arrhythmic hamsters causes memory deficits in object recognition, spatial memory, and contextual fear conditioning. This project will bridge the fields of circadian rhythms and learning and memory, and provide functional evidence for an SCN output pathway for cognition. The main strength of this application is that our manipulations should restore memory rather than induce deficits. Neuroanatomical studies have shown that the SCN innervates the MS of rats, mice, and golden hamsters, thus, we firmly believe that the functional relationship between the SCN and MS established in Siberian hamsters will generalize to other species. The arrhythmic hamster is an excellent model of circadian dysfunction in humans because the light treatment leaves animals neurologically and genetically intact. Thus, a major benefit of this proposal is that the results will lead to a much needed animal model in which we can study circadian contributions to human pathologies that are of interest to clinicians.